Grok 4.5 outperforms competitors in AI coding benchmarks
Crypto Briefing 2026-07-17 07:17:03
Context: Grok 4.5, an AI coding model, has achieved a top ranking in the SWE Marathon benchmark, surpassing competitors like Claude Opus 4.8 and Fable. The model's performance was announced in a recent report. Grok 4.5's success in the benchmark is notable for its high accuracy and cost-effectiveness.
Key Facts
- Grok 4.5 leads the SWE Marathon benchmark with a performance score of 29.0%, outpacing its closest competitors, Claude Opus 4.8 and Fable.
- The AI model's API pricing is set at $2 per million token, making it a competitively priced option for users.
- Grok 4.5's achievement in the benchmark demonstrates its capabilities in AI coding, showcasing its potential for developers and organizations seeking efficient coding solutions.